Volume 18, Issue 1 (5-2021)
JSDP 2021, 18(1): 28-13 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Tahmasbi H, Jalali M, Shakeri H. A social recommender system based on matrix factorization considering dynamics of user preferences. JSDP 2021; 18 (1) :28-13
URL: http://jsdp.rcisp.ac.ir/article-1-929-en.html
URL: http://jsdp.rcisp.ac.ir/article-1-929-en.html
Islamic Azad university,Mashhad Branch
Abstract: (2118 Views)
With the expansion of social networks, the use of recommender systems in these networks has attracted considerable attention. Recommender systems have become an important tool for alleviating the information that overload problem of users by providing personalized recommendations to a user who might like based on past preferences or observed behavior about one or various items. In these systems, the users’ behavior is dynamic and their preferences change over time for different reasons. The adaptability of recommender systems to capture the evolving user preferences, which are changing constantly, is essential.
Recent studies point out that the modeling and capturing the dynamics of user preferences lead to significant improvements in recommendation accuracy. In spite of the importance of this issue, only a few approaches recently proposed that take into account the dynamic behavior of the users in making recommendations. Most of these approaches are based on the matrix factorization scheme. However, most of them assume that the preference dynamics are homogeneous for all users, whereas the changes in user preferences may be individual and the time change pattern for each user differs. In addition, because the amount of numerical ratings dramatically reduced in a specific time period, the sparsity problem in these approaches is more intense. Exploiting social information such as the trust relations between users besides the users’ rating data can help to alleviate the sparsity problem. Although social information is also very sparse, especially in a time period, it is complementary to rating information. Some works use tensor factorization to capture user preference dynamics. Despite the success of these works, the processing and solving the tensor decomposition is hard and usually leads to very high computing costs in practice, especially when the tensor is large and sparse.
In this paper, considering that user preferences change individually over time, and based on the intuition that social influence can affect the users’ preferences in a recommender system, a social recommender system is proposed. In this system, the users’ rating information and social trust information are jointly factorized based on a matrix factorization scheme. Based on this scheme, each users and items is characterized by a sets of features indicating latent factors of the users and items in the system. In addition, it is assumed that user preferences change smoothly, and the user preferences in the current time period depend on his/her preferences in the previous time period. Therefore, the user dynamics are modeled into this framework by learning a transition matrix of user preferences between two consecutive time periods for each individual user. The complexity analysis implies that this system can be scaled to large datasets with millions of users and items. Moreover, the experimental results on a dataset from a popular product review website, Epinions, show that the proposed system performs better than competitive methods in terms of MAE and RMSE.
Recent studies point out that the modeling and capturing the dynamics of user preferences lead to significant improvements in recommendation accuracy. In spite of the importance of this issue, only a few approaches recently proposed that take into account the dynamic behavior of the users in making recommendations. Most of these approaches are based on the matrix factorization scheme. However, most of them assume that the preference dynamics are homogeneous for all users, whereas the changes in user preferences may be individual and the time change pattern for each user differs. In addition, because the amount of numerical ratings dramatically reduced in a specific time period, the sparsity problem in these approaches is more intense. Exploiting social information such as the trust relations between users besides the users’ rating data can help to alleviate the sparsity problem. Although social information is also very sparse, especially in a time period, it is complementary to rating information. Some works use tensor factorization to capture user preference dynamics. Despite the success of these works, the processing and solving the tensor decomposition is hard and usually leads to very high computing costs in practice, especially when the tensor is large and sparse.
In this paper, considering that user preferences change individually over time, and based on the intuition that social influence can affect the users’ preferences in a recommender system, a social recommender system is proposed. In this system, the users’ rating information and social trust information are jointly factorized based on a matrix factorization scheme. Based on this scheme, each users and items is characterized by a sets of features indicating latent factors of the users and items in the system. In addition, it is assumed that user preferences change smoothly, and the user preferences in the current time period depend on his/her preferences in the previous time period. Therefore, the user dynamics are modeled into this framework by learning a transition matrix of user preferences between two consecutive time periods for each individual user. The complexity analysis implies that this system can be scaled to large datasets with millions of users and items. Moreover, the experimental results on a dataset from a popular product review website, Epinions, show that the proposed system performs better than competitive methods in terms of MAE and RMSE.
Keywords: Social recommender system, rating prediction, preference dynamics, matrix factorization, trust
Type of Study: Research |
Subject:
Paper
Received: 2018/11/13 | Accepted: 2020/08/18 | Published: 2021/05/22 | ePublished: 2021/05/22
Received: 2018/11/13 | Accepted: 2020/08/18 | Published: 2021/05/22 | ePublished: 2021/05/22
References
1. [1] M. Hosseini, M. Nasrollahi, and A. Baghaei. "A hybrid recommender system using trust and bi-clustering in order to increase the efficiency of collaborative filtering". JSDP, vol 15, no. 2, pp.119-132, 2018. [DOI:10.29252/jsdp.15.2.119]
2. [2] A. Y. Aravkin, K. R. Varshney, and L. Yang, "Dynamic matrix factorization with social influence," in 2016 IEEE International Workshop on Machine Learning for Signal Processing, 2016, pp. 1-6. [DOI:10.1109/MLSP.2016.7738846]
3. [3] H. Bao, Q. Li, S. S. Liao, S. Song, and H. Gao, "A new temporal and social PMF-based method to predict users' interests in micro-blogging," Decision Support Systems, vol. 55, no. 3, pp. 698-709, 2013. [DOI:10.1016/j.dss.2013.02.007]
4. [4] I. Barjasteh, R. Forsati, D. Ross, A. H. Esfahanian, and H. Radha, "Cold-start recommendation with provable Guarantees: A decoupled approach," IEEE Transactions on Knowledge and Data Engineering, vol. 28, no. 6, pp. 1462-1474, 2016. [DOI:10.1109/TKDE.2016.2522422]
5. [5] J. Cheng, Y. Liu, H. Zhang, X. Wu, and F. Chen, "A new recommendation algorithm based on user's dynamic information in complex social network," Mathematical Problems in Engineering, vol. 2015, pp. 1-6, 2015. [DOI:10.1155/2015/281629]
6. [6] W. Cheng, G. Yin, Y. Dong, H. Dong, and W. Zhang, "Collaborative filtering recommenda-tion on users' interest sequences," Plos One, vol. 11, no. 5, p. e0155739, 2016. [DOI:10.1371/journal.pone.0155739] [PMID] [PMCID]
7. [7] W.-S. Chin, B.-W. Yuan, M.-Y. Yang, Y. Zhuang, Y.-C. Juan, and C.-J. Lin, "LIBMF: a library for parallel matrix factorization in shared-memory systems," The Journal of Machine Learning Research, vol. 17, no. 1, pp. 2971-2975, 2016.
8. [8] E. Frolov and I. Oseledets, "Tensor methods and recommender systems," Wiley Inter-disciplinary Reviews: Data Mining and Knowledge Discovery, vol. 7, no. 3, pp. 1-41, 2017. [DOI:10.1002/widm.1201]
9. [9] J. Gaillard, "Recommender systems : Dynamic adaptation and argumentation.", Ph.D. disser-tation, Dept. Computer Science, Avignon Univ., Avignon, France, 2014.
10. [10] G. Guo, J. Zhang, and N. Yorke-Smith, "A novel recommendation model regularized with user trust and item ratings," IEEE Transactions on Knowledge and Data Engineering, vol. 28, no. 7, pp. 1607-1620, 2016. [DOI:10.1109/TKDE.2016.2528249]
11. [11] B. Ju, Y. Qian, M. Ye, R. Ni, and C. Zhu, "Using dynamic multi-task non-negative matrix factorization to detect the evolution of user preferences in collaborative filtering," PLoS ONE, vol. 10, no. 8, pp. 1-20, 2015. [DOI:10.1371/journal.pone.0135090] [PMID] [PMCID]
12. [12] Y. Koren, "Collaborative filtering with temporal dynamics," Communications of the ACM, vol. 53, no. 4, pp. 89-97, 2010. [DOI:10.1145/1721654.1721677]
13. [13] S. Li and Y. Fu, "Robust representations for response prediction," in Robust Repre-sentation for Data Analytics, Springer, pp. 147-174, 2017. [DOI:10.1007/978-3-319-60176-2_8]
14. [14] D. R. Liu, K. Y. Chen, Y. C. Chou, and J. H. Lee, "Online recommendations based on dynamic adjustment of recommendation lists," Knowledge-Based Systems, vol. 161, pp. 375-389, 2018. [DOI:10.1016/j.knosys.2018.07.038]
15. [15] Y. Y. Lo, W. Liao, C. S. Chang, and Y. C. Lee, "Temporal matrix factorization for tracking concept drift in individual user preferences," IEEE Transactions on Computational Social Systems, vol. 5, no. 1, pp. 156-168, 2018. [DOI:10.1109/TCSS.2017.2772295]
16. [16] W. Lu, S. Ioannidis, S. Bhagat, and L. V. S. Lakshmanan, "Optimal recommendations under attraction, aversion, and social influence," Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '14, 2014, pp. 811-820. [DOI:10.1145/2623330.2623744] [PMCID]
17. [17] C. Luo, X. Cai, and N. Chowdhury, "Self-training temporal dynamic collaborative filtering," In Proceedings of the 18th Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer, 2014, pp. 461-472. [DOI:10.1007/978-3-319-06608-0_38]
18. [18] A. Mnih and R. Salakhutdinov, "Probabilistic matrix factorization," in Advances in Neural Information Processing Systems. Cambridge, MA, USA: MIT Press, 2008, pp. 1257-1264.
19. [19] J. Pan, Z. Ma, Y. Pang, and Y. Yuan, "Robust probabilistic tensor analysis for time-variant collaborative filtering," Neurocomputing, vol. 119, pp. 139-143, 2013. [DOI:10.1016/j.neucom.2012.03.035]
20. [20] F. S. F. Pereira, J. Gama, S. de Amo, and G. M. B. Oliveira, "On analyzing user preference dynamics with temporal social networks," Machine Learning, vol. 107, no. 11, pp. 1745-1773, 2018. [DOI:10.1007/s10994-018-5740-2]
21. [21] D. Rafailidis, "Modeling trust and distrust information in recommender systems via joint matrix factorization with signed graphs," in Proceedings of the 31st Annual ACM Symposium on Applied Computing, 2016, pp. 1060-1065. [DOI:10.1145/2851613.2851697]
22. [22] D. Rafailidis, P. Kefalas, and Y. Manolopoulos, "Preference dynamics with multimodal user-item interactions in social media recommendation," Expert Systems with Applications, vol. 74, pp. 11-18, 2017. [DOI:10.1016/j.eswa.2017.01.005]
23. [23] D. Rafailidis and A. Nanopoulos, "Modeling users preference dynamics and side information in recommender systems," IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 46, no. 6, pp. 782-792, 2016. [DOI:10.1109/TSMC.2015.2460691]
24. [24] C. Rana and S. K. Jain, "An evolutionary clustering algorithm based on temporal features for dynamic recommender systems," Swarm and Evolutionary Computation, vol. 14, pp. 21-30, 2014. [DOI:10.1016/j.swevo.2013.08.003]
25. [25] N. Sahoo, D. A. Tepper, and T. Mukhopadhyay, "A hidden markov model for collaborative filtering," MIS Quarterly, vol. 36, no. 4, pp. 1329-1356, 2012. [DOI:10.2307/41703509]
26. [26] Y. Shi, M. Larson, and A. Hanjalic, "Collaborative filtering beyond the user-item matrix : A survey of the state of the art and future challenges," ACM Computing Surveys (CSUR), vol. 47, no. 1, p. 3, 2014. [DOI:10.1145/2556270]
27. [27] A. P. Singh and G. J. Gordon, "Relational learning via collective matrix factorization," Proceeding of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD 08, 2008, pp. 650-658. [DOI:10.1145/1401890.1401969] [PMID]
28. [28] J. Z. Sun, D. Parthasarathy, and K. R. Varshney, "Collaborative kalman filtering for dynamic matrix factorization," IEEE Transactions on Signal Processing, vol. 62, no. 14, pp. 3499-3509, 2014. [DOI:10.1109/TSP.2014.2326618]
29. [29] H. Tahmasbi, M. Jalali, and H. Shakeri, "Modeling Temporal Dynamics of User Preferences in Movie Recommendation," in 2018 8th International Conference on Computer and Knowledge Engineering (ICCKE), 2018, pp. 194-199. [DOI:10.1109/ICCKE.2018.8566316]
30. [30] J. Tang, "Epinions Dataset." [Online]. Available: http://www.cse.msu.edu/~tang-jili/trust.html. [Accessed: 05-Jan-2018].
31. [31] J. Tang, H. Gao, A. Das Sarma, Y. Bi, and H. Liu, "Trust evolution: Modeling and its applications," IEEE Transactions on Knowledge and Data Engineering, vol. 27, no. 6, pp. 1724-1738, 2015. [DOI:10.1109/TKDE.2014.2382576]
32. [32] C. Tong, J. Qi, Y. Lian, J. Niu, and J. J. P. C. Rodrigues, "TimeTrustSVD: A collaborative filtering model integrating time, trust and rating information," Future Generation Computer Systems, vol. 93, pp. 933-941, 2019. [DOI:10.1016/j.future.2017.07.037]
33. [33] J. Wang, S. Zhang, X. Liu, Y. Jiang, and M. Zhang, "A novel collective matrix factorization model for recommendation with fine-grained social trust prediction," Con-currency Computation, vol. 29, no. 19, pp. 1-14, 2017. [DOI:10.1002/cpe.4233]
34. [34] H. Wu, K. Yue, Y. Pei, B. Li, Y. Zhao, and F. Dong, "Collaborative topic regression with social trust ensemble for recommendation in social media systems," Knowledge-Based Systems, vol. 97, pp. 111-122, 2016. [DOI:10.1016/j.knosys.2016.01.011]
35. [35] T. Wu, Y. Feng, J. Sang, B. Qiang, and Y. Wang, "A novel recommendation algorithm incorporating temporal dynamics, reviews and item correlation," IEICE TRANS-ACTIONS on Information and Systems, vol. 101, no. 8, pp. 2027-2034, 2018. [DOI:10.1587/transinf.2017EDP7387]
36. [36] L. Xiong, X. Chen, T.-K. Huang, J. Schneider, and J. G. Carbonell, "Temporal collaborative filtering with bayesian probabilistic tensor factorization," in Proceedings of the 2010 SIAM International Conference on Data Mining, 2010, pp. 211-222. [DOI:10.1137/1.9781611972801.19] [PMCID]
37. [37] B. Yang, Y. Lei, J. Liu, and W. Li, "Social collaborative filtering by trust," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 8, pp. 1633-1647, Aug. 2017. [DOI:10.1109/TPAMI.2016.2605085] [PMID]
38. [38] X. Yang, Y. Guo, Y. Liu, and H. Steck, "A survey of collaborative filtering based social recommender systems," Computer Communi-cations, vol. 41, pp. 1-10, 2014. [DOI:10.1016/j.comcom.2013.06.009]
39. [39] H. Yin, B. Cui, L. Chen, Z. Hu, and Z. Huang, "A temporal context-aware model for user behavior modeling in social media systems," in Proceedings of the 2014 ACM SIGMOD international conference on Management of data, 2014, no. 1, pp. 1543-1554. [DOI:10.1145/2588555.2593685] [PMID] [PMCID]
40. [40] Y. Yu, Y. Gao, H. Wang, and R. Wang, "Joint user knowledge and matrix factorization for recommender systems," World Wide Web, vol. 21, no. 4, pp. 1141-1163, 2018. [DOI:10.1007/s11280-017-0476-7]
41. [41] F. Yu, A. Zeng, S. Gillard, and M. Medo, "Network-based recommendation algorithms: A review," Physica A: Statistical Mechanics and its Applications, vol. 452, pp. 192-208, 2016. [DOI:10.1016/j.physa.2016.02.021]
42. [42] Z. Zhang and H. Liu, "Social recommendation model combining trust propagation and sequential behaviors," Applied Intelligence, vol. 43, no. 3, pp. 695-706, 2015. [DOI:10.1007/s10489-015-0681-y]
43. [43] F. Zhang, Q. Liu, and A. Zeng, "Timeliness in recommender systems," Expert Systems with Applications, vol. 85, pp. 270-278, 2017. [DOI:10.1016/j.eswa.2017.05.038]
44. [44] C. Zhang, K. Wang, H. Yu, J. Sun, and E.-P. Lim, "Latent factor transition for dynamic collaborative filtering," in Proceedings of the 2014 SIAM International Conference on Data Mining, 2014, pp. 452-460. [DOI:10.1137/1.9781611973440.52]
45. [45] W. Zhang, Y. Du, T. Yoshida, and Y. Yang, "DeepRec: A Deep Neural Network Approach to Recommendation with Item Embedding and Weighted Loss Function," Information Sciences, vol. 470, pp. 121-140, 2019. [DOI:10.1016/j.ins.2018.08.039]
46. [46] B. Zou, C. Li, L. Tan, and H. Chen, "GPUTENSOR: efficient tensor factorization for context-aware recommendations," Infor-mation Sciences, vol. 299, pp. 159-177, 2015. [DOI:10.1016/j.ins.2014.12.004]
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
